Related Experiment Video
Updated: May 28, 2026
14:14
Targeting Neuronal Fiber Tracts for Deep Brain Stimulation Therapy Using Interactive, Patient-Specific Models
Published on: August 12, 2018
Artificial Intelligence for STN-DBS Surgical Planning in Parkinson's Disease: A Multicenter Study Comparing
Feifei Wu1, Raffaella Buonanno2, Valentina Baro1
1Academic Neurosurgery, Department of Neuroscience, University of Padua, 35128 Padua, Italy.
Brain Sciences
|May 27, 2026
Summary
Artificial intelligence (AI) shows comparable results to traditional methods for determining Deep Brain Stimulation (DBS) targets in Parkinson's disease (PD) patients, particularly for lateral-lateral coordinates. Further research is needed to validate AI's role in surgical planning.
Area of Science:
- Neurosurgery
- Medical Imaging
- Artificial Intelligence
Background:
- Deep Brain Stimulation (DBS) is an established therapy for advanced Parkinson's disease (PD).
- Accurate target identification is crucial for effective DBS surgery.
- Traditional methods for target identification include direct and indirect approaches.
Purpose of the Study:
- To compare the efficacy of traditional targeting methods with artificial intelligence (AI) for DBS in Parkinson's disease.
- To evaluate the accuracy of AI-driven target identification against conventional atlases and intraoperative recordings.
Main Methods:
- Analysis of eight patients undergoing bilateral subthalamic nucleus (STN) DBS.
- Comparison of target coordinates derived from Schaltenbrand and Wahren atlases, AI (RebrAIn system), and microelectrode recordings (MERs).
- Statistical evaluation using non-parametric ANOVA Friedman test for stereotactic coordinate differences (X, Y, Z).
Main Results:
- Significant agreement observed in lateral-lateral (X) coordinates across all methods.
- Substantial discrepancies noted in antero-posterior (Y) and cranio-caudal (Z) coordinates.
- AI demonstrated comparable lateral-lateral (X) coordinate values to traditional methods.
Conclusions:
- AI shows promise in DBS target determination, exhibiting comparable accuracy to traditional methods along the X-axis.
- Interindividual anatomical variability and imaging limitations present challenges.
- Further validation of AI and machine learning models is essential for integrating them into preoperative workflows.
